Abstract:
This thesis explores the integration of Internet of Things (IoT) technology and computer vision to create an advanced student behavior monitoring system in Rwandan schools, aiming to automate attendance tracking and enhance discipline monitoring. The system employs computer vision and edge processing for real-time attendance analysis and face recognition, utilizing ESP32 CAM, and android device camera for efficient data capture and processing. Parents receive immediate SMS notifications for absences and monthly behavior reports via respective email, fostering improved communication. The edge processing capabilities allow for rapid detection and response to abnormal activities, ensuring a safe learning environment through swift staff intervention. By leveraging machine learning techniques, the system effectively identifies and records unusual activities, such as sudden noise, and promptly alerts authorized personnel to remotely take any necessary actions. Remote monitoring tools enable staff to oversee classroom activities, promoting a secure educational setting. This system surpasses manual methods by automating comprehensive report generation, thus reducing the need for constant human intervention. Additionally, it addresses the limitations of webbased systems by enhancing parental involvement through real-time updates and detailed emailed reports. The edge processing features provide daily updates and customizable configurations, ensuring a user-friendly experience tailored to the needs of academic and professional institutions. Overall, the system offers improved attendance accuracy, stringent discipline monitoring, personalized child support, early parent-child engagement, and early detection of abnormal behavior.
